amazon-ads-budget-pacing
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources (Amazon Ads campaigns). Ingestion points: campaign data and performance actuals are read into context (SKILL.md). Boundary markers: instructions specify that approval is bound to a unique proposal ID and account context, requiring user confirmation (SKILL.md). Capability inventory: the skill utilizes local Python scripts (allocate.py, report.py) for data processing and calls Ads MCP tools for mutations (SKILL.md). Sanitization: the allocate.py script performs comprehensive schema and value validation on the input data bundle before any logic is applied.
Audit Metadata